1. Field of the Invention
The present invention relates to a method of manufacturing a semiconductor memory device, and more particularly, to a method of manufacturing storage nodes of a semiconductor memory device.
2. Description of the Related Art
As semiconductor memory devices become more highly integrated, the sizes, structures and methods of fabricating the devices are continually changing. For example, a method of separating devices using trenches is widely used, and the sizes and the structures of components such as gate electrodes, contacts and capacitors have changed. In addition, capacitors have changed from two-dimensional structures into three-dimensional structures. Here, due to the increase in the integration of semiconductor memory devices, the areas of the capacitors have been reduced while the heights of the capacitors have been increased.
With the area of capacitors being continuingly reduced, various methods have been suggested to secure the large capacitance required by the semiconductor memory devices. A prime example of these methods is increasing the volume of the capacitors. In order to increase the volume of the capacitors, three-dimensional capacitors are formed or a layer of hemispherical grains is formed on the surfaces of capacitor electrodes. Further, increasing the height of storage nodes has also been suggested.
The thickness of a sacrificial oxide layer, such as a mold oxide layer used to form the storage nodes, is increased due to the increase in the height of the storage nodes. In addition to the problem of depositing the sacrificial oxide layer in a thick and uniform manner, there is also a problem with removing the sacrificial oxide layer after the storage nodes have been manufactured. Removing the sacrificial oxide layer was not a serious problem when the heights of the storage nodes were not very high. However, as the heights of the storage nodes increase, removing the sacrificial oxide layer becomes more problematic.
FIG. 1 is a flowchart illustrating a conventional method of manufacturing storage nodes of a semiconductor memory device. In the method of FIG. 1, an etch stopping layer and a mold oxide layer are sequentially deposited on a substrate or an interlevel insulating layer to form storage nodes in step S11. Thereafter, the mold oxide layer is patterned to define a region where the storage nodes are to be formed and exposed portions of the etch stopping layer are removed in step S12. Next, a conductive layer is deposited on a storage node region and the mold oxide layer to a predetermined thickness in step S13. A buffer oxide layer is thickly deposited on the conductive layer in step S14. The buffer oxide layer and the conductive layer are etched using chemical mechanical polishing (“CMP”) or a dry etchback method to separate nodes of the conductive layer in step S15, thus completing the storage nodes.
Thereafter, the mold oxide layer and the buffer oxide layer, which surround the conductive layer, are removed altogether in step S16. Here, the mold oxide layer and the buffer oxide layer are removed in a batch type by wet etching. Wet etching is performed using a buffered oxide etchant (“BOE”) solution or a hydrogen fluoride (“HF”) solution. After silicon oxide layers are removed by wet etching, a rinse and drying process is performed in step S17.
In the case where the thick layers are removed by wet etching due to the increase in the heights of the storage nodes, bridges are likely to occur between adjacent storage nodes. In other words, the storage nodes can fall onto adjacent storage nodes, and short circuits can occur. One reason why the storage nodes fall is because of the presence of water between the surfaces of adjacent storage nodes. Here, water remains between storage nodes after a wet etching process or a rinse process.
Bridges of the storage nodes due to the presence of water will now be described with reference to FIG. 2. As shown in FIG. 2, when a water film 20 forms between two storage nodes 16a, two forces are present and exerted on the storage nodes 16a. Here, the two forces include surface tension Fs, which tries to pull adjacent storage nodes 16a, and a shear and bending force Fe, which acts in the opposite direction to the surface tension Fs.
It is assumed that cylindrical storage nodes are formed in a rectangular parallelepiped structure while being firmly attached to a substrate 10 as rigid beams. In this case, the shear and bending force Fe can be represented by Equation 1.
                              F          e                =                              3            ·            E            ·            I            ·            x                                H            3                                              (        1        )            
Here, E denotes Young's modulus, I denotes the moment of inertia of a horizontal cross section, H denotes the height of the storage nodes, and x denotes a deformation distance.
The surface tension Fs can be represented by Equation 2.Fs=2·v·sin θ·(L+H)  (2)
Here, v denotes the surface tension coefficient of water, θ denotes the contact angle between a storage node and water, and L denotes the distance of one side of the storage node.
At equilibrium, the forces have the same magnitudes. Thus, the deformation distance can be represented as shown in Equation 3 by combining Equations 1 and 2.
                    x        =                                            2              ·              v              ·              sin                        ⁢                                                  ⁢                          θ              ·                              (                                  L                  +                  H                                )                            ·                              H                3                                                          3            ·            E            ·            I                                              (        3        )            
Referring to Equation 3, the deformation distance x is proportional to the height H of the storage nodes.
On the other hand, as represented by Equation 4, the possibility of the storage nodes falling is proportional to the deformation distance x and inversely proportional to the distance D between the storage nodes. Here, since the deformation distance x is proportional to the height H of the storage nodes, the possibility of the storage nodes falling is proportional to the height H of the storage nodes and inversely proportional to the distance D between the storage nodes. Accordingly, as the design rule is reduced and the height of the capacitors is increased, the possibility of the storage nodes falling is increased.
                              possibility  of  adhesion                ∝                                                                              deformation  distance  of                                                                                                                          storage  node                                      ⁢                                      (                    x                    )                                                                                                                                            distance  between                                                                                                                          storage  nodes                                      ⁢                                      (                    D                    )                                                                                                          (        4        )            
Turning to FIG. 3, the reference numeral 300 generally indicates a scanning electron microscope (“SEM”) photograph illustrating the fallen state of two storage nodes 312 and 314 in the case where the height of the storage nodes is about 18,000 Å and the distance between the storage nodes in a row direction is about 700 Å. Here, the photograph is taken from above the storage nodes 310, 312, 314 and 316. The SEM photograph of FIG. 3 is taken after the oxide layers, i.e., the mold oxide layer and the buffer oxide layer, have been removed by the conventional BOE solution and the rinse and drying processes have been performed. The circle denotes the two fallen storage nodes.
Typically, a dryer is used for the drying process after the oxide layers have been removed by wet etching. The likelihood of storage nodes falling generally varies according to the type of dryer used and its performance. Since the design rule is continuingly decreased, it is generally desirable to decrease the distances between the storage nodes and to increase the heights of the storage nodes. Since the performance of the dryer is ultimately limited, there is a corresponding limitation in the ability to prevent the storage nodes from falling merely by improving the performance of the dryer. Accordingly, what is needed is an improvement over the conventional wet etching method that does not depend on the performance of a dryer.